huggingface/transformers

Pull Requests Commits

push encoder changes

ita.zaporozhets@huggingface.co committed 360 days ago

ca88d25f

adding blt local decoder / encoder models

itazap committed 364 days ago

bc151480

itazap committed 1 year ago

55f0da70

ita.zaporozhets@huggingface.co committed 1 year ago

5268deb7

[FIX] Save speed metrics to logs (#38136)

pavelgein committed 1 year ago

Verified 0f77ca72

Omit creation of positional IDs within ESM if applicable (#38089)

simonlevine committed 1 year ago

Verified 27ef46e8

disable deepspeed when setting up fake trainer (#38101)

winglian committed 1 year ago

Verified fe9426f1

enable trainer test cases on xpu (#38138)

yao-matrix committed 1 year ago

Verified 7caa57e8

Hotfix: Flash Attention 2 support in Pixtral (#38146)

uminaty committed 1 year ago

Verified b11b28cc

[generate] Run custom generation code from the Hub (#36405)

gante committed 1 year ago

Verified 0e0e5c10

Remove head mask in generative models (#35786)

zucchini-nlp committed 1 year ago

Verified 955e61b0

enable csm integration cases on xpu, all passed (#38140)

yao-matrix committed 1 year ago

Verified 0173a99e

[Qwen3] Qwen3 MoE add tp plan for expert mlps (#38135)

hgt312 committed 1 year ago

Verified e5a48785

Fix incorrect attention mask truncate in WhisperFlashAttention2 (#36477)

OliBomby committed 1 year ago

Verified 4005e30c

enable d_fine finetuning properly (#37962)

SangbumChoi committed 1 year ago

Verified aa27fa75

Add `manueldeprada` to `run_slow` whitelist (#38126)

manueldeprada committed 1 year ago

Verified e021bf6b

[docs] add uv installation instructions for source builds (#37968)

arjunaskykok committed 1 year ago

Verified ef27b2bc

Update trainer.md (#38113)

guspuffygit committed 1 year ago

Verified 4a2decd1

Add config validation and style tweaks (#37589)

Kirire committed 1 year ago

Verified 935bbbc7

Fix auto batch size finder test (#38125)

ivarflakstad committed 1 year ago

Verified 1b009663

Fix temporal padding in Qwen2VLImageProcessor when the number of frames is not divisible by temporal_patch_size (#38076)

ritwickchaudhry committed 1 year ago

Verified fe918d13

[video processor] fix tests (#38104)

zucchini-nlp committed 1 year ago

Verified aaf224d5

enable finegrained_fp8 and granite_speech cases on XPU (#38036)

yao-matrix committed 1 year ago

Verified 9b5ce556

Fix description and formatting errors in code docs (#38074)

bilibili12433014 committed 1 year ago

Verified b311a3f5

Add style bot (#38102)

SunMarc committed 1 year ago

Verified b499a14b

[CSM] update test for t4 runners (#38110)

eustlb committed 1 year ago

Verified e0f225cb

Add Fast Image Processor for vilt (#37304)

devxaitist committed 1 year ago

Verified 342961f6

Fix InternVL interpolate_pos_encoding and add to video_processing_auto (#38092)

yonigozlan committed 1 year ago

Verified 8771766a

fix `check_bad commit.py` gives wrong results (#38107)

ydshieh committed 1 year ago

Verified 582d5e0e

[bug] fix llava processor to calculate unpadding size correctly (#37988)

cyr0930 committed 1 year ago

Verified a5cc7a67

Older